W5: Data Wrangling with Tidy Data, Part 2

Modifying and creating new columns

The mutate() function takes in the following arguments: the first argument is the dataframe of interest, and the second argument is a new or existing data variable that is defined in terms of other data variables.

I think of them like formulas in Excel. . . .

metadata |>
  select(ModelID, Age) |>
  head()

metadata2 <- metadata |>
              mutate(olderAge = Age + 10) 

metadata2 |>
  select(ModelID, Age, olderAge) |>
  head()

metadata |>
  mutate(olderAge = Age + 10)

expression data.frame

expression$KRAS_Exp[1:10]

 [1] 4.634012 4.638653 4.032101 5.503031 3.713696 3.972693 3.235727 4.135042
 [9] 9.017365 3.940167

expression2 = expression |>
  mutate(log_KRAS_Exp = log(KRAS_Exp))

expression2$log_KRAS_Exp[1:10]

 [1] 1.533423 1.534424 1.394288 1.705299 1.312028 1.379444 1.174254 1.419498
 [9] 2.199152 1.371223

Try it Out 1

One thing we often need to do is scale or normalize data. Let’s try scaling by the maximum value in a column. Try calculating a new variable KRAS_scaled by calculating:

KRAS_Exp / max(KRAS_Exp)

Try it Out 2

Try calcluating a new variable in mutation called TP53_CDKN2A which is equal to:

• TP53_Mut | CDKN2A_Mut

Then filter new_mutation by TP53_CDKN2A == TRUE. How many rows did you return?

Grouping and summarizing dataframes

metadata_grouped = group_by(metadata, OncotreeLineage)
metadata_type = summarise(metadata_grouped, MeanAge = mean(Age, na.rm = TRUE), Count = n())

head(metadata_type)

Grouping and summarizing dataframes

The group_by() function returns the identical input dataframe but remembers which variable(s) have been marked as grouped.

Try it Out 1

First group by OncotreeLineage. Then try calculating the maximum age as max_age within metadata when it is grouped by OncotreeLineage:

Quick Note

Our summarization has a lot of NAs. We can use the na.rm argument in max() to drop NA values in our calculation.

Try it Out 2

First group by OncotreeLineage. Then try calculating the median age as median_age within metadata when it is grouped by OncotreeLineage. (Hint: use the median() function. Bonus points if you use the na.rm argument)

Think about this:

What is the difference between these two pipelines?

arrange()

If we need to sort by a column, we can use arrange():

metadata |>
  arrange(Age)

Descending order

If we want to sort by descending order, then we can wrap our column in desc() (short for descending):

metadata |>
  arrange(desc(Age))

Sorting by Multiple Columns

We can sort by multiple columns by passing in multiple variables:

metadata |>
  arrange(desc(Age), ModelID)

Try it Out

Try sorting by desc(Age) and OncotreeLineage, and then by OncotreeLineage and desc(Age). Does order matter?

Merging two dataframes together

Merging two dataframes together

One strategy we can use is to only merge on common ids between the two tables. If one table has ids that aren’t in the second table, then we’ll remove those rows.

We see that in both dataframes, the rows (observations) represent cell lines with a common column ModelID, so let’s merge these two dataframes together, using inner_join():

merged = inner_join(metadata, expression, by = "ModelID")

dim(metadata)

[1] 1864   30

dim(expression)

[1] 1450  536

dim(merged)

[1] 1450  565

Which rows do we keep from each table?

inner_join() keeps all observations common to both dataframes based on the common column defined via the by argument.

merged |> 
  select(ModelID, KRAS_Exp, OncotreeLineage)

Merging two dataframes together: variations

Try it Out

Try out inner_join() on the two tables and compare the number of rows in the merged table.

Now try full_join() on the two tables and compare the number of rows in the merged table.

Next Week

Learning Community Week!

Suggest topics here: